3

It seems unintuitive to me to use italic text in theorem and lemma text styles, because some mathematical symbols (letters) are almost identical to regular italicised letters.

Consider the following example (text sourced from this website). I got the wonderful newtheoremstyle solution from this SE thread, from a user named Mico. Thank you, Mico!

\documentclass{article}

\usepackage[utf8]{inputenc}
\usepackage[english]{babel}
\usepackage{amsthm}

\newtheoremstyle{bettertheorem}
{}                % Space above
{}                % Space below
{\slshape}        % Theorem body font % (default is "\upshape")
{}                % Indent amount
{\bfseries}       % Theorem head font % (default is \mdseries)
{.}               % Punctuation after theorem head % default: no punctuation
{ }               % Space after theorem head
{}                % Theorem head spec

\begin{document}

\newtheorem{theorem}{Theorem}

\begin{theorem}
Given two line segments whose lengths are $a$ and $b$ respectively there is a 
real number $r$ such that $b=ra$.
\end{theorem}

\theoremstyle{bettertheorem}
\newtheorem{bettertheorem}{Theorem}

\begin{bettertheorem}
Given two line segments whose lengths are $a$ and $b$ respectively there is a 
real number $r$ such that $b=ra$.
\end{bettertheorem}

\centering
\huge{\textbf{Which one looks better to \emph{you}?}}

\end{document}

enter image description here

Now, I can see how this question could get flagged as "primarily opinion-based", so let me rephrase my question as objectively as possible: Is there any rational reason to use italic style instead of slanted in theorem texts, considering that italicised text and math text look more alike (and are, at least for me, harder to distinguish and read) than slanted text and math text?

Thank you for your answers in advance and God bless.

God bless
  • 1,767
  • To answer the question in the image, undisputagly, the italic shape… – Bernard Jul 29 '19 at 07:55
  • Tradition? Ugliness of slanted type? – egreg Jul 29 '19 at 07:55
  • I like the slanted style better, it's easier to read because it is more like normal text. I don't know what the reason is for using italics but if I would have to guess it is probably just for historical reasons (italic fonts being more easily available or something to that effect). – Marijn Jul 29 '19 at 08:02
  • 2
    @egreg I noticed that people often call \slshape ugly on this SE. While it's not the prettiest of styles, I don't see it as particularly ugly. Even if \itshape were objectively prettier, I think that the visual trade-off is worth it, as it is easier to distinguish math symbols from text when using \slshape. Not to mention the danger of confusing indefinite article 'a' with some math symbol a. – God bless Jul 29 '19 at 08:03
  • @Bernard I really have no idea why so many people find slanted style ugly, man... – God bless Jul 29 '19 at 08:03
  • @Marijn My thoughts exactly! Do you know of any math textbook or journal which is beginning to incorporate slanted text instead of italic? – God bless Jul 29 '19 at 08:05
  • @GregorPerčič I don't like slanted type (except for sans serif); you like it. Go in a library and try to find books using slanted serif. – egreg Jul 29 '19 at 08:14
  • 1
    @egreg While I do like slanted type better, this is not my primary concern. As I've said, the foremost reason to switch is, in my opinion, the confusion the italic style produces with its proximate similarity to letters in math mode, hence the potential confusion of the indefinite article 'a' with a math symbol a, as I've pointed out. I don't know, I just think that it would be easier to read. – God bless Jul 29 '19 at 08:18
  • I am font shape agnostic, but If distinguish math mode is the main concern, then slanted font seems to me nearly as bad as the italic font. I cannot say if that is a sacrilege for the math book traditions, but an upright sans serif font will be clearly distinguished from both math mode and main serif text. It is often used in captions to avoid confusion with main text , so why not also in theorems? – Fran Jul 29 '19 at 08:53
  • 1
    I'm a font agnostic for the most part, and I must confess that I can't get too excited by this discussion. For all serif fonts I'm familiar with which provide both slanted-italic ("italic") and slanted-roman ("slanted") font faces, the only truly-meaningful differences pertain to just three lowercase letters: a, e, and f. (For most sans-serif fonts, what's commonly called "italic" should actually be called "slanted".) FWIW, the TeXbook uses slanted-roman, not italics, for emphasis, and this feature has certainly never either excited or bothered me. – Mico Jul 29 '19 at 09:13
  • 3
    Basically, this is a matter of history and tradition. Although I rather like slanted roman for various reasons, this seems to me to be a matter of opinion, and therefore off topic. But Knuth did recognize the ambiguity of single math letters like "a" -- Computer Modern has a different font for math, which is wider than the CM text italic. This is a distinction not generally supported by other font families now used for math. – barbara beeton Jul 29 '19 at 15:19
  • @barbarabeeton: Indeed, and it is a rather bad history and tradition. I don't even know why on earth definitions should be upright and theorems should be italic, when both contain mathematical statements, which the main text also contains! Even Knuth himself used upright style for the Properties he listed for what he felt would describe aesthetically pleasing curve-fitting in Mathematical Typography, which are no different from Axioms. – user21820 Oct 08 '19 at 18:06
  • Even if the purpose is emphasis, the italic Computer Modern Font is really quite hard to read compared to almost any other font I've seen. – user21820 Oct 08 '19 at 18:07

0 Answers0